2024-02-26_10-14-02
This model is a fine-tuned version of openai/whisper-small.en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2555
- Wer: 9.2582
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 64
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2454 | 1.0 | 711 | 0.2441 | 9.3061 |
| 0.0952 | 2.0 | 1422 | 0.2452 | 9.4815 |
| 0.0664 | 3.0 | 2133 | 0.2555 | 9.2582 |
Framework versions
- Transformers 4.37.2
- Pytorch 1.13.1+cu116
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 82
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for ThomasFG/E3-67.5-67.5
Base model
openai/whisper-small.en